In both developed and emerging-economies, major cities continue to experience increasing traffic congestion. To address this issue, complex Traffic Management Systems (TMS) are employed in recent years to help manage traffic. These systems fuse traffic-surveillance-related information from a variety of sensors deployed across traffic networks. A TMS requires real-time information to make effective control decisions and to deliver trustworthy information to users, such as travel time, congestion level, etc. There are three fundamental inputs required by TMS, namely, traffic volume, vehicular speed, and traffic density. Using conventional traffic loop detectors one can directly measure flow and velocity. However, traffic density is more difficult to measure. The situation becomes more difficult for multi-lane motorways due to drivers lane-change behaviour. This research investigates statistical modelling and analysis of traffic flow. It contributes to the literature of transportation and traffic management and research in several aspects. First, it takes into account lane-changes in traffic modelling through incorporating a Markov chain model to describe the drivers lane-change behaviour. Secondly, the lane change probabilities between two adjacent lanes are not assumed to be fixed but rather they depend on the current traffic condition. A discrete choice model is used to capture drivers lane choice behaviour. The drivers choice probabilities are modelled by several traffic-condition related attributes such as vehicle time headway, traffic density and speed. This results in a highly nonlinear state equation for traffic density. To address the issue of high nonlinearity of the state space model, the EKF and UKF is used to estimate the traffic density recursively. In addition, a new transformation approach has been proposed to transform the observation equation from a nonlinear form to a linear one so that the potential approximation in the EKF & UKF can be avoided. Numerical studies have been conducted to investigate the performance of the developed method. The proposed method outperformed the existing methods for traffic density estimation in simulation studies. Furthermore, it is shown that the computational cost for updating the estimate of traffic densities for a multi-lane motorway is kept at a minimum so that online applications are feasible in practice. Consequently the traffic densities can be monitored and the relevant information can be fed into the traffic management system of interest.

Šis darbas skirtas Italijos verslininkų pasitenkinimo darbu analizei, naudojantis kokybinių požymių daugiamatės statistikos metodais. Duomenys yra iš Italijos Veneto regiono verslininkų tyrimo, kurį 2006 metais pradėjo Padujos universiteto Statistikos departamentas. Imtį sudaro 1216 stebėjimų iš beveik 113000 Veneto regiono (Šiaurės rytų Italija) verslininkų populiacijos. Priklausomasis kintamasis, t.y. pasitenkinimas darbu, išmatuotas ranginėje keturių taškų Likert skalėje su šiomis kategorijomis: „nepatenkintas“, „nei patenkintas, nei nepatenkintas“, „pakankamai patenkintas“ ir „patenkintas“. Paaiškinamuosius kintamuosius sudaro demografinės, firmos charakteristikos, tapimo verslininku motyvus, darbą, laisvalaikį ir ateities perspektyvas atspindintys kintamieji. Beveik visi kintamieji, taip pat ir priklausomasis, yra kategoriniai. Todėl pagrindinis šio darbo tikslas yra išsirinkti geriausiai tokio tipo duomenims tinkantį modelį. Aptariama galimybė naudotis tiesine regresija, atlikus optimalųjį matavimo skalės transformavimą (optimal scaling), logistine regresija, prieš tai apjungus priklausomojo kintamojo rangus, daugianariu logit modeliu, analizuojant kiekvienos kategorijos tikimybę atskirai ir rangine regresija su logit ir cloglog sąryšio funkcijomis. Potencialių pasitenkinimo darbu veiksnių skaičius- pakankamai didelis (daugiau negu 30), todėl itin svarbus uždavinys yra parinkti aiškinančiuosius kintamuosius, vengiant multikolinearumo problemos. Siekiant sumažinti... [toliau žr. visą tekstą] / The paper presents job satisfaction analysis among Italian entrepreneurs using multivariate statistical techniques. Data are taken from Veneto region (North-East part of Italy) businessmen research started by Department of Statistics of Padova University at 2006. Sample consists of 1216 observations (real population is almost 113 000 entrepreneurs at this region). The outcome variable for the job satisfaction is measured on an ordered, categorical, four-point Likert scale – „dissatisfied‘“, „neither satisfied nor dissatisfied“ , „quite satisfied“ and „satisfied“. Explanatory variables include demographic items, firm characteristics, variables representing the reasons of having started the own business, items associated with work, leisure activities and future perspectives. Quite all variables, as well as the dependent one, beeing categorical, the main objective of this work is to select an appropriate model for such type of data among possible alternatives. Discussions are made on possibility to applicate linear regression with optimal scaling, binary logit for dichotomized dependent variable, multinomial logit for analysis of every single category and ordinal logistic regression with several link functions (logit and cloglog). The number of possible determinants of job satisfaction beeing quite large (there were more than 30 questionnaire items associated with job satisfaction) it was of a great importance choosing which explanatory variables should be included in the models... [to full text]

The dynamics around managing urban infrastructure are changing dramatically. Today???s infrastructure management challenges ??? in the wake of shrinking coffers and stricter stakeholders??? requirements ??? include finding better condition assessment tools and prediction models, and effective and intelligent use of hard-earn data to ensure the sustainability of urban infrastructure systems. Wastewater collection networks ??? an important and critical component of urban infrastructure ??? have been neglected, and as a result, municipalities in North America and other parts of the world have accrued significant liabilities and infrastructure deficits. To reduce cost of ownership, to cope with heighten accountability, and to provide reliable and sustainable service, these systems need to be managed in an effective and intelligent manner. The overall objective of this research is to present a new strategic management framework and related tools to support multi-perspective maintenance, rehabilitation and replacement (M, R&R) planning for wastewater collection networks. The principal objectives of this research include: (1) Developing a comprehensive wastewater collection network asset database consisting of high quality condition assessment data to support the work presented in this thesis, as well as, the future research in this area. (2) Proposing a framework and related system to aggregate heterogeneous data from municipal wastewater collection networks to develop better understanding of their historical and future performance. (3) Developing statistical models to understand the deterioration of wastewater pipelines. (4) To investigate how strategic management principles and theories can be applied to effectively manage wastewater collection networks, and propose a new management framework and related system. (5) Demonstrating the application of strategic management framework and economic principles along with the proposed deterioration model to develop long-term financial sustainability plans for wastewater collection networks. A relational database application, WatBAMS (Waterloo Buried Asset Management System), consisting of high quality data from the City of Niagara Falls wastewater collection system is developed. The wastewater pipelines??? inspections were completed using a relatively new Side Scanner and Evaluation Technology camera that has advantages over the traditional Closed Circuit Television cameras. Appropriate quality assurance and quality control procedures were developed and adopted to capture, store and analyze the condition assessment data. To aggregate heterogeneous data from municipal wastewater collection systems, a data integration framework based on data warehousing approach is proposed. A prototype application, BAMS (Buried Asset Management System), based on XML technologies and specifications shows implementation of the proposed framework. Using wastewater pipelines condition assessment data from the City of Niagara Falls wastewater collection network, the limitations of ordinary and binary logistic regression methodologies for deterioration modeling of wastewater pipelines are demonstrated. Two new empirical models based on ordinal regression modeling technique are proposed. A new multi-perspective ??? that is, operational/technical, social/political, regulatory, and finance ??? strategic management framework based on modified balanced-scorecard model is developed. The proposed framework is based on the findings of the first Canadian National Asset Management workshop held in Hamilton, Ontario in 2007. The application of balanced-scorecard model along with additional management tools, such as strategy maps, dashboard reports and business intelligence applications, is presented using data from the City of Niagara Falls. Using economic principles and example management scenarios, application of Monte Carlo simulation technique along with the proposed deterioration model is presented to forecast financial requirements for long-term M, R&R plans for wastewater collection networks. A myriad of asset management systems and frameworks were found for transportation infrastructure. However, to date few efforts have been concentrated on understanding the performance behaviour of wastewater collection systems, and developing effective and intelligent M, R&R strategies. Incomplete inventories, and scarcity and poor quality of existing datasets on wastewater collection systems were found to be critical and limiting issues in conducting research in this field. It was found that the existing deterioration models either violated model assumptions or assumptions could not be verified due to limited and questionable quality data. The degradation of Reinforced Concrete pipes was found to be affected by age, whereas, for Vitrified Clay pipes, the degradation was not age dependent. The results of financial simulation model show that the City of Niagara Falls can save millions of dollars, in the long-term, by following a pro-active M, R&R strategy. The work presented in this thesis provides an insight into how an effective and intelligent management system can be developed for wastewater collection networks. The proposed framework and related system will lead to the sustainability of wastewater collection networks and assist municipal public works departments to proactively manage their wastewater collection networks.

This thesis investigates how a tax and warning label on less healthy snack food products may affect consumer behaviour when the imposition of the tax is a source of consumer information. A survey that included choice experiments was implemented in supermarkets. Participants were asked to choose between high fat snacks, some displaying a stigmatizing warning label, and healthier snacks. Multinomial logit and latent class models exploring choice were estimated and a predictive hypothetical market was set up. Results show that the warning label had a negative price premium of about $4. The effect of price, though small, becomes even smaller as BMI increases. A fat tax for health is not recommended because it might not hit the target population, people were not very price sensitive, and it would likely be regressive. To encourage health, it appears to be more effective to display a warning label than to apply a tax. / Agricultural and Resource Economics

本研究主要是應用二元分量迴歸BQR（Binary Quantile Regression）模型的方法估計銀行對中小企業授信之信用評等，以期提早偵測出可能會有違約還款的企業，達到授信時的預警效果。信用評等目的為協助金融機構在貸放前更明確的瞭解企業的信用風險，並具以衡量是否核准貸款的重要依據。在過去的研究中最廣為應用的計量方法主要為有母數（parametric）區別迴歸模型，包括Logit Model和Probit Model等區別迴歸模型，這二種模型在正確的條件設定之下，模型的預測結果可以說相當的好，但若是估計資料的分配並未符合所設定的條件，或者是資料具有無法觀察到的異質變異（heteroskedastic），則估計結果會有顯著的偏誤。傳統區別模型的一般設定如下，假設發生違約的機率給定為： ，此處 表示實際上是否真的發生違約逾期還款的情形。 為了在估計時更能控制風險，最近許多有關信用評等的研究方法傾向使用半無母數（semiparametric）單一指數模型以及無母數（nonparametric）的估計方法，如類神經網路與歸納樹（classification trees）分析方法。 而本文主要是將半無母數的分量迴歸區別模型和過去以有母數為主的Probit及Logit區別迴歸模型做比較。Koenker和Bassett（1978）提出分量迴歸估計方法（Quantile Regression Methods），分量迴歸可以更完整的反應出共變異效果對被解釋變數的影響，除此之外，分量迴歸模式提供使用上較多的彈性，在估計時無需對母體的分配做假設，另外，和傳統的最小平方（OLS）估計法不同在於OLS給予估計參數的分量為50%，因此OLS估計出的迴歸線只有一條，因此分析解釋變數對被解釋變數的影響是平均效果；分量迴歸區別模型則給予估計參數不同百分比的分量，從而可在相同樣本下得到不同的分量迴歸線，觀察解釋變數對於被解釋變數影響程度的變化，因此藉由不同分量估計出不同的迴歸係數 ，可以更加瞭解整體分配的全貌。

En la actualidad uno de los grandes desafíos relacionados con el transporte en las urbes consiste en frenar el constante aumento del uso del automóvil reemplazándolo por el de modos de transporte más eficientes como el transporte público. De acuerdo con lo anterior, es importante conocer qué caracteriza y motiva la constante migración de usuarios de transporte público hacia el automóvil. El entendimiento de cómo influyen las distintas características de la oferta de transporte público en los distintos segmentos de la población, permite medir qué grupos están en mayor riesgo y tomar medidas preventivas para evitar el abandono. En la literatura relacionada con el transporte son pocos los estudios que se enfocan en entender el abandono y la forma de retener a los usuarios de transporte público. Debido a esto fue necesario realizar una revisión bibliográfica amplia, incluyendo tanto temas relacionados con transporte como otros relacionados con fuga de clientes enfocados al marketing y a las finanzas, para así conocer de mejor manera las hipótesis y herramientas usadas para abordar el tema en general. Al revisar distintas formas de modelar, se optó finalmente por utilizar como base un modelo logit multinomial de partición modal del tipo jerárquico. A dicho modelo, se le realizaron algunas modificaciones para captar de mejor forma los factores determinantes en el abandono, entre estos destaca la inclusión de una función que recopila las características más importantes que determinan el abandono o la fidelidad con el transporte público. Es importante destacar que además del modelo teórico, fue implementada una aplicación computacional con la cuál fue posible validar las hipótesis planteadas a partir de una base de datos. Para la aplicación se utiliza la base de datos Panel de Santiago, que permite establecer el seguimiento de una muestra de usuarios (trabajadores-punta mañana) en distintos cortes temporales, antes y después de la implementación del sistema de transporte público de Santiago (Transantiago), el cual produjo profundos cambios en la oferta. El análisis de esta base de datos permite extraer hipótesis e identificar de las características de los usuarios que migran desde el transporte público. Los datos del panel junto con otras fuentes de información complementarias de calidad de servicio y nivel de satisfacción de usuarios, sirvieron como datos de entrada para calibrar el modelo econométrico propuesto en esta tesis específicamente para predecir el abandono, el cual fue construido a partir de las hipótesis extraídas del Panel de Santiago sobre el abandono. Los resultados del modelo mostraron que el ingreso, la calidad de servicio, el grupo etario, las actividades extras al trabajo, entre otros factores, son relevantes a la hora de conocer qué usuarios están en riesgo de abandonar. Con algunas de estas variables se construyeron escenarios para analizar la sensibilidad del modelo y predecir el abandono en cortes temporales futuros.

Nesta tese são desenvolvidos três estudos sobre as reservas internacionais brasileiras, utilizando diferentes técnicas econométricas, com o objetivo de determinar a influência de medidas absolutas e relativas de reservas sobre o rating soberano de crédito e o spread soberano, bem como o nível adequado para garantir a liquidez externa. As análises foram feitas com dados mensais do período jan/2000-jun/2008. No primeiro estudo, mostrou-se que diferentes medidas de reservas internacionais apresentam efeito significativo na explicação do rating soberano de crédito, através de modelos ordered logit para a média dos ratings emitidos pelas três principais agências (Moody's, Standard & Poors e Fitch). Entretanto, o indicador de maior poder explicativo não foi o nível absoluto de reservas, mas a razão entre dívida pública externa líquida e PIB. Outras variáveis de destacada importância na maioria dos modelos foram o percentual da dívida interna de curto prazo, investimento estrangeiro direto/PIB e inflação. Variáveis tradicionalmente utilizadas como indicadores de liquidez, como razão reservas/importações e conta corrente/PIB, não foram significativas na maioria dos modelos. Os resultados confirmam os indícios contidos no discurso das agências de rating, quanto à importância das reservas internacionais em sua avaliação, embora alertando que outras variáveis, como perfil de endividamento do governo e perspectivas de crescimento, são também fundamentais. O segundo estudo de caso encontrou relação significativa entre as reservas internacionais e o spread soberano, através de modelos de correção de erros. O efeito estimado do rating soberano foi não-significativo ou pouco explicativo, comparado aos fundamentos, provavelmente devido à volatilidade do spread soberano em resposta a variações nas condições do mercado, ao contrário do rating. O melhor modelo obtido utilizou o nível absoluto de reservas, evidenciando também efeitos significativos da aversão global ao risco, taxas de juros internacionais e crises políticas internas. Os resultados desse estudo indicam custo marginal decrescente das reservas internacionais e a necessidade de considerá-lo endógeno em modelos de minimização de custos para determinação do nível ótimo de reservas. O terceiro estudo implementou a metodologia de Liquidity-at-Risk sugerida por Greenspan (1999) para avaliar a adequação do nível de reservas internacionais para a manutenção da liquidez externa. Para a medida de liquidez reservas/dívida externa de curto prazo (razão de Guidotti), estimou-se que o nível de reservas internacionais mantidas pelo Brasil em jun/2008 (US$200 bilhões) era aproximadamente o dobro do necessário para garantir uma razão de Guidotti superior a 1, com 99% de probabilidade, durante 24, 36 ou 48 meses. Em diversos cenários alternativos de percentual das dívidas externa e interna de curto prazo, meta de superávit primário, índice de aversão ao risco e taxas de juros externas, as reservas iniciais necessárias situaram-se em US$85-105 bilhões. A análise de custos revela que o aumento das reservas diminui os juros médios da dívida, embora efeito maior pudesse ser alcançado através do aumento do superávit primário. As evidências sugerem que a motivação das autoridades brasileiras para a manutenção de reservas em torno de US$200 bilhões não é puramente precaucionária, admitindo as hipóteses de ganho de credibilidade e flexibilidade para a execução da política fiscal. / This thesis developed three case studies on the Brazilian international reserves, using various econometric techniques in order to determine the influence of absolute and relative measures of reserves over both the sovereign credit rating and the sovereign spread, as well as to assess the adequate reserves level to ensure external liquidity. Analyses were carried out on monthly data from Jan/2000 to Jun/2008. The first case study found significant effects of different reserves measures in explaining the sovereign credit rating, by fitting ordered logit models to the average of the ratings issued by the three main agencies (Moody's, Standard & Poors and Fitch) for the Brazilian long term external debt. However, the best explaining variable was not the absolute level of reserves, but the ratio "net public external debt/GDP" instead. It was noteworthy the significance of the following variables in most of the models tested: short term internal debt (%), foreign direct investment/GDP and inflation. Variables traditionally used as external liquidity measures, like reserves/imports and current account/GDP, are not statistically significant in most of the models fitted in this study. Results support the evidence found in the rating agencies' reports, as to the importance of international reserves in their credit quality assessment, although pointing to other variables, like government debt profile and growth perspectives, as equally critical. The second case study found significant relationship between the Brazilian international reserves and its sovereign spread, using error correction models. The estimated effect of sovereign rating was either non-significant, or poorly explanatory when compared to macroeconomic fundamentals, probably due to the volatility of sovereign spread in response to changes in market conditions, unlike the sovereign rating. The best model obtained included the absolute level of reserves, showing also significant effect of the global risk aversion, external interest rates and internal political crises. The results of this study point to a decreasing marginal cost of international reserves and the need of considering it as endogenous in optimal reserves models based in cost minimization. Finally, the third case study implemented the Liquidity-at-Risk methodology suggested by Greenspan (1999), in order to assess the Brazilian reserves level adequacy in maintaining external liquidity. For the liquidity measure adopted - the ratio "reserves/short term external debt" (Guidotti's ratio) - it was found that the Brazilian reserves level held in Jun/2008 (US$200 billion) was roughly twice the necessary one to ensure a Guidotti's ratio above 1, with 99% probability, within 24, 36 or 48 months. In several alternative scenarios varying the short term external debt, short term internal debt, primary surplus, global risk aversion and external interest rates, the required initial reserves was in the range US$85-105 billion. An analysis of alternative policies' costs revealed the expected effect of higher reserves in decreasing the average debt service, although a dramatically higher impact would be obtained by an increase in primary surplus. Evidence suggest that the Brazilian authorities motivation for holding international reserves as high as US$200 billion may not be purely precautionary, pointing to the hypotheses of credibility gains and fiscal flexibility issues.

Submitted by Cassius Figueiredo (cassius.figueiredo@fgvmail.br) on 2015-10-02T14:14:49Z No. of bitstreams: 1 Dissertação_Cassius_Figueiredo.pdf: 1419517 bytes, checksum: 0bc08a99261a6abda6c7592ca62cddaa (MD5) / Approved for entry into archive by Janete de Oliveira Feitosa (janete.feitosa@fgv.br) on 2015-10-06T13:08:24Z (GMT) No. of bitstreams: 1 Dissertação_Cassius_Figueiredo.pdf: 1419517 bytes, checksum: 0bc08a99261a6abda6c7592ca62cddaa (MD5) / Approved for entry into archive by Marcia Bacha (marcia.bacha@fgv.br) on 2015-10-13T13:12:20Z (GMT) No. of bitstreams: 1 Dissertação_Cassius_Figueiredo.pdf: 1419517 bytes, checksum: 0bc08a99261a6abda6c7592ca62cddaa (MD5) / Made available in DSpace on 2015-10-13T13:12:33Z (GMT). No. of bitstreams: 1 Dissertação_Cassius_Figueiredo.pdf: 1419517 bytes, checksum: 0bc08a99261a6abda6c7592ca62cddaa (MD5) Previous issue date: 2015-08-31 / A common scenario in countries with inflation targeting regimes is Central Bank intervention in exchange market to keep volatility under control. Those interventions are commom in developing countries. In Brazil interventions are mostly done through spot market, derivatives market (exchange rate swaps) and also through forward market operations, liquidity and borrowing lines. Due to low volumes related to the last three, we kept our efforts concentrated in interventions through spot and derivatives markets. There are several articles discussing how successful those interventions are but only a few evaluating which factors may lead Central Bank of Brazil to an intervention. We try to fill this gap using two different techniques: the ever present logistic regression and a new approach (as fas as we know) using artificial neural networks. In parallel we will try to define if there are specific factors affecting different scenarios of intervention. The dataset goes from 2005 to 2012, period where Central Bank of Brazil has been intervening based on market demand and not in a standardized way for longer periods of time. Our results show that there may be factors more relevant to one of the intervention decisions (buying or selling dollars) and we can highlight the relevance of exchange rate volatility, particularly in interventions where Central Bank of Brazil is buying dollars. This result is fully aligned with other papers on the subject. / Em economias com regimes de metas de inflação é comum que Bancos Centrais intervenham para reduzir os níveis de volatilidade do dólar, sendo estas intervenções mais comuns em países não desenvolvidos. No caso do Brasil, estas intervenções acontecem diretamente no mercado à vista, via mercado de derivativos (através de swaps cambiais) ou ainda com operações a termo, linhas de liquidez e via empréstimos. Neste trabalho mantemos o foco nas intervenções no mercado à vista e de derivativos pois estas representam o maior volume financeiro relacionado à este tipo de atuação oficial. Existem diversos trabalhos que avaliam o impacto das intervenções e seus graus de sucesso ou fracasso mas relativamente poucos que abordam o que levaria o Banco Central do Brasil (BCB) a intervir no mercado. Tentamos preencher esta lacuna avaliando as variáveis que podem se relacionar às intervenções do BCB no mercado de câmbio e adicionalmente verificando se essas variáveis se relacionam diferentemente com as intervenções de venda e compra de dólares. Para tal, além de utilizarmos regressões logísticas, como na maioria dos trabalhos sobre o tema, empregamos também a técnica de redes neurais, até onde sabemos inédita para o assunto. O período de estudo vai de 2005 a 2012, onde o BCB interveio no mercado de câmbio sob demanda e não de forma continuada por longos períodos de tempo, como nos anos mais recentes. Os resultados indicam que algumas variáveis são mais relevantes para o processo de intervenção vendendo ou comprando dólares, com destaque para a volatilidade implícita do câmbio nas intervenções que envolvem venda de dólares, resultado este alinhado com outros trabalhos sobre o tema.

Submitted by Cristiane Oliveira (cristiane.oliveira@fgv.br) on 2016-03-16T20:24:22Z No. of bitstreams: 1 Dissertacao_Versao Final_pos Banca_v3.pdf: 1185057 bytes, checksum: feed2dbb84423bff900a25bdb3589ead (MD5) / Approved for entry into archive by Renata de Souza Nascimento (renata.souza@fgv.br) on 2016-03-16T21:21:17Z (GMT) No. of bitstreams: 1 Dissertacao_Versao Final_pos Banca_v3.pdf: 1185057 bytes, checksum: feed2dbb84423bff900a25bdb3589ead (MD5) / Made available in DSpace on 2016-03-16T21:21:39Z (GMT). No. of bitstreams: 1 Dissertacao_Versao Final_pos Banca_v3.pdf: 1185057 bytes, checksum: feed2dbb84423bff900a25bdb3589ead (MD5) Previous issue date: 2016 / The intention of this work is to study how the level of capital as a percentage of riskweighted assets, i.e. the Basel Ratio, maintained by the financial intermediaries can serve as a predictor of the failure of these financial intermediaries. One of the challenges was the lack of availability of the Basel Ratio for each institution before 2009, while most of bank failures in Brazil occurred in the period between 1995 and 2005. Considering this, a Synthetic Basel Ratio (IBS) for the period from December 1995 to December 2014 was created, testing the assumption that a higher level of capital in relation to risk assets reduces the likelihood of such a failure. Both logit models with a discrete binary variable and a survival analysis are used, which allow an estimate to be made of how much of an increase in the level of capital brings in terms of life for the institution. The sample studied consist of 313 publicly and privately-owned financial intermediaries operating in Brazil, using semi-annually figures. In line with previous studies, empirical evidence was found pointing to an inverse relation between the level of capital and the likelihood of a failure, seen in the use of both the logit and survival models. / Este trabalho tem como proposta estudar se o nível de capital sobre os ativos ponderados pelo risco, o Índice de Basileia, mantido pelos intermediários financeiros, pode servir como preditor de falência dos intermediários financeiros. Um dos desafios apresentados foi o fato do Índice de Basileia reportado para cada instituição estar disponível ao público apenas a partir de 2009, ao passo que grande parte das falências bancárias no Brasil ocorreram no período entre 1995 e 2005. Dessa forma, construindo um Índice de Basileia Sintético (IBS) para o período de dezembro de 1995 a dezembro de 2014, testou-se a hipótese de que um nível mais alto de capital em relação aos ativos de risco diminui a probabilidade de falência da instituição. São utilizados modelos logit com variável binária discreta e análise survival, possibilitando estimar o quanto que um aumento no nível de capital proporciona em tempo de vida para a instituição. A amostra estudada é composta por 313 intermediários financeiros atuando no Brasil, tanto de controle público quanto privado, com dados semestrais. Em linha com estudos anteriores, foi encontrada evidência empírica apontando para uma relação inversa entre nível de capital e probabilidade de falência, tanto com o emprego de logit como de survival.

Made available in DSpace on 2010-04-20T20:48:13Z (GMT). No. of bitstreams: 3 86614.pdf.jpg: 20961 bytes, checksum: 6b5baa8cf53a88a3b8e4b93c17c6cf58 (MD5) 86614.pdf.txt: 333427 bytes, checksum: 7e0721735b5ba8c069a7f3dcbc54abee (MD5) 86614.pdf: 1853538 bytes, checksum: d8d6acc9301b160b1b7cbf06ec75b279 (MD5) Previous issue date: 2003-05-16T00:00:00Z / The main objective of this dissertation is to estimate the joint price-elasticity of demand, decomposed in brand choice and quantity price-elasticity, and discuss the implications of this decomposition for a specific product category, using a scanner data sample of Brazilian households. Eight hypotheses were tested through two models. The first one refers to the rand choice decision, and a conditional logit model based on household utility maximization was used. The second one encompassed demand equations, modeled through classical linear regressions. Both models were specified to test the dependence of the two purchase decisions. Regarding to the validity, the brand choice model demonstrated a satisfactory prevision power, compared to similar studies in the literature. Managerial implications include specific price decisions to each brand, as the nature of the price-elasticity decomposition varies among brands. / O objetivo geral desta tese é estimar a elasticidade-preço da demanda de forma conjunta, decomposta em elasticidade-preço da escolha da marca e da quantidade comprada, e discutir as implicações desta decomposição para uma categoria específica de produto. Para isto foram usados dados escaneados de uma amostra de domicílios no contexto varejista brasileiro. Oito hipóteses foram testadas por meio de dois modelos. O primeiro refere-se à decisão de escolha da marca, em que foi empregado o modelo logit condicional baseado na maximização da utilidade do domicílio. O segundo envolveu equações de demanda, obtidas pelo modelo clássico de regressão linear. Ambos foram especificados de forma que se pudesse testar a dependência das duas decisões de compra. No que diz respeito à validação, o modelo de escolha da marca demonstrou uma satisfatória capacidade de previsão, comparativamente aos modelos analisados na literatura. Implicações gerenciais incluem específicas decisões e ações de preço para as marcas, já que a natureza da decomposição das elasticidades-preço varia entre marcas.